DOI: https://doi.org/10.2151/jmsj.2024-020
Development of a Temperature Prediction Method Combining Deep Neural Networks and a Kalman Filter
##article.authors##
-
Takuya Inoue
Numerical Prediction Development Center, Japan Meteorological Agency
https://researchmap.jp/jma_inoue
-
Tsuyoshi Thomas Sekiyama
Meteorological Research Institute, Japan Meteorological Agency
https://researchmap.jp/thomas_sekiyama
- Atsushi Kudo Numerical Prediction Division, Japan Meteorological Agency
DOI:
https://doi.org/10.51094/jxiv.505Keywords:
deep convolutional neural network, statistical post-processing, temperature forecast, Kalman filter, fine-tuningAbstract
Numerical weather forecast models have biases caused by insufficient grid resolution and incomplete physical processes, especially near the land surface. Therefore, the Japan Meteorological Agency (JMA) has been operationally post-processing the forecast model outputs to correct the biases. The operational post-processing method uses a Kalman filter (KF) algorithm for surface temperature prediction. Recent reports showed that deep convolutional neural networks (CNNs) were better than the JMA operational method in correcting the temperature forecast biases. This study combined the CNN-based bias correction scheme with the JMA operational KF algorithm. We expected that the combination of CNNs and a KF would improve the post-processing performance, as the CNNs modify large horizontal structures, and then, the KF corrects minor spatiotemporal deviations. As expected, we confirmed that the combination outperformed both CNNs and the KF alone. This study demonstrates the advantages of the new method in correcting coastal front, heat wave, and radiative cooling biases.
Conflicts of Interest Disclosure
The authors declare that they have no conflict of interest.Downloads *Displays the aggregated results up to the previous day.
References
Bing, G., M. Langguth, Y. Ji, A. Mozaffari, S. Stadtler, K. Mache, and M. G. Schultz, 2022: Temperature forecasting by deep learning methods. Geosci. Model Dev., 15, 8931–8956, doi:10.5194/gmd-15-8931-2022.
Dongjin, C., Y. Cheolhee, S. Bokyung, I. Jungho, Y. Donghyuck, and C. Dong-Hyun, 2022: A novel ensemble learning for post-processing of NWP Model's next-day maximum air temperature forecast in summer using deep learning and statistical approaches. Wea. Climate Extremes, 35, doi:10.1016/j.wace.2022.100410.
Furuichi, Y., and N. Matsuzawa 2009: Snowfall amount guidance. Textbook for Numerical Weather Prediction. 47, Japan Meteorological Agency, 27-38 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwptext/47/chapter2.pdf.]
Glahn, B., K. Gilbert, R. Cosgrove, D. P. Ruth, and K. Sheets, 2009: The gridding of MOS. Wea. Forecasting, 24, 520-529, doi:10.1175/2008WAF2007080.1.
Glahn, H. R., D. A. Lowry, 1972: The Use of Model Output Statistics (MOS) in Objective Weather Forecasting, J. Appl. Meteor. Climatol., 11, 1203-1211, doi:10.1175/1520-0450(1972)011<1203:TUOMOS>2.0.CO;2.
Hara, T., 2014: Studies on recent remarkable cases. Textbook for Numerical Weather Prediction. 47, Japan Meteorological Agency, 118-144 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwptext/47/chapter4.pdf.]
Hara, T., and H. Kurahashi, 2017: Changes in the characteristics of meso-scale numerical prediction system. Textbook for Numerical Weather Prediction. 50, Japan Meteorological Agency, 48-55 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwptext/50/chapter2.pdf.]
Japan Meteorological Agency, 1986: Progress of guidance in Japan. Textbook for Weather Prediction Technique, 35, Japan Meteorological Agency, 19-20 (in Japanese)Japan Meteorological Agency, 2021: Development results. Numerical Prediction Development Center Annual Report 2020. Japan Meteorological Agency, 15-76 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/npdc/r02/npdc_annual_report_r02_2-02.pdf.]
Japan Meteorological Agency, 2022: Meso-scale numerical prediction system vertical layer augmentation, forecast range extension, and physical process improvements. Numerical Prediction Development Center Annual Report 2021. Japan Meteorological Agency, 92-99 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/npdc/r03/npdc_annual_report_r03_4-03.pdf.]
Japan Meteorological Agency, 2023a: NWP application products. Outline of the Operational Numerical Weather Prediction at the Japan Meteorological Agency. Japan Meteorological Agency, 157-188. [Available at https://www.jma.go.jp/jma/jma-eng/jma-center/nwp/outline-latest-nwp/pdf/outline2023_04.pdf.]
Japan Meteorological Agency, 2023b: Data Assimilation Systems. Outline of the Operational Numerical Weather Prediction at the Japan Meteorological Agency. Japan Meteorological Agency, 14 pp [Available at https://www.jma.go.jp/jma/jma-eng/jma-center/nwp/outline-latest-nwp/pdf/outline2023_02.pdf.]
Japan Meteorological Agency, 2023c: Numerical Weather Prediction Models. Outline of the Operational Numerical Weather Prediction at the Japan Meteorological Agency. Japan Meteorological Agency, 53-156 [Available at https://www.jma.go.jp/jma/jma-eng/jma-center/nwp/outline-latest-nwp/pdf/outline2023_03.pdf.]
Kawano, K., M. Ujiie, M. Kunii, and S. Nishimoto, 2019: Meso-scale ensemble prediction system. Textbook for Numerical Weather Prediction. 52, Japan Meteorological Agency, 1-15 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwptext/52/chapter1.pdf.]
Kingma, D. P., and J. L. Ba, 2015: Adam: A method for stochastic optimization. Conference paper at the Third International Conference on Learning Representations 2015, San Diego, U.S.A., 15 pp, doi:10.48550/arXiv.1412.6980.
Klein, W. H., and H. R. Glahn, 1974: Forecasting local weather by means of Model Output Statistics. Bull. Amer. Meteor. Soc., 55, 1217-1227, doi:10.1175/1520-0477(1974)055<1217:FLWBMO>2.0.CO;2.
Kudo 2022: Statistical Post-Processing for Gridded Temperature Prediction Using Encoder–Decoder-Based Deep Convolutional Neural Networks, J. Meteorol. Soc. Japan., 100, 2019-232, doi:10.2151/jmsj.2022-011.
Kuroki, Y., 2017: Improvement of gridded temperature guidance and changes of guidance for snowfall amount and categorized weather. Textbook for Numerical Weather Prediction. 50, Japan Meteorological Agency, 94-101 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwptext/50/chapter4.pdf.]
Kusabiraki, H., and S. Moriyasu, 2013: Verification in the operational numerical weather prediction models. Report of Numerical Prediction Division. 59, Japan Meteorological Agency, 16-24 (in Japanese).
Kusabiraki, H., 2020: Radiation. Report of Numerical Prediction Division. 66, Japan Meteorological Agency, 61-68 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwpreport/66/chapter2.pdf.]
Météo-France, 2015: JOINT WMO TECHNICAL PROGRESS REPORT ON THE GLOBAL DATA PROCESSING AND FORECASTING SYSTEM AND NUMERICAL WEATHER PREDICTION RESEARCH ACTIVITIES FOR 2015, Météo-France. [Available at https://wmoomm.sharepoint.com/:w:/s/wmocpdb/EfAO2BxjA7NDgdRscNaL488Brn1XlsY87wefjuIq7uRBug.]
Météo-France, 2020: JOINT WMO TECHNICAL PROGRESS REPORT ON THE GLOBAL DATA PROCESSING AND FORECASTING SYSTEM AND NUMERICAL WEATHER PREDICTION RESEARCH ACTIVITIES FOR 2015, Météo-France. [Available at https://wmoomm.sharepoint.com/:b:/s/wmocpdb/EQOKo5TMgDRAjyv3SFawEqIBGl4OKHS4txHR_PufRbIWTA.]
Met Office, 2015: JOINT WMO TECHNICAL PROGRESS REPORT ON THE GLOBAL DATA PROCESSING AND FORECASTING SYSTEM AND NUMERICAL WEATHER PREDICTION RESEARCH ACTIVITIES FOR 2015, Met Office. [Available at https://wmoomm.sharepoint.com/:b:/s/wmocpdb/EfYms7k874hOry4vAkDu-MoBOVC9363_VDOrvxJM-Sar8g.]
Miura, H., and Y. Ohashi, 2017: Influences of the relocations of the Okayama Local Meteorological Observatory on the measured air-temperature. Naturalistae, 21, 7-15.
Nair, V., and G. E. Hinton, 2010: Rectified linear units improve restricted Boltzmann machines. Proceedings of the Twenty-seventh International Conference on Machine Learning (ICML-10), Haifa, Israel, 807-814.
Sannohe, Y., 2018: Temperature guidance. Report of Numerical Prediction Division. 64, Japan Meteorological Agency, 132-143 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwpreport/64/chapter4.pdf.]
Segami, T. M. Obayashi, M. Kunitsugu, and T. Fujita, 1995: Kalman filter. Textbook for Numerical Weather Prediction, 28, Japan Meteorological Agency, 66-78 (in Japanese)
Sheridan, P., S. Smith, A. Brown, and S. Vosper, 2010: A simple height-based correction for temperature downscaling in complex terrain. Meteor. Appl., 17, 329-339, doi:10.1002/met.177.
Shi, X., Z. Chen, H. Wang, D. Yeung, W. Wong, W. Woo, 2015: Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting. Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015, NeurIPS 2015, 7-12 December 2015, Montreal, Quebec, Canada. Cortes, C., N. D. Lawrence, D. D. Lee, M. Sugiyama, and R. Garnett (eds.), 802-810, doi:10.48550/arXiv.1506.04214.
Suzuki, K., T. Iwasaki, and T. Yamazaki, 2021: Analysis of systematic error in Numerical Weather Prediction of coastal fronts in Japan's Kanto Plain. J. Meteor. Soc. Japan, 99, 27-47, doi:10.2151/jmsj.2021-002.
Takada, S., 2018a: Introduction to guidance. Report of Numerical Prediction Division. 64, Japan Meteorological Agency, 3-8 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwpreport/64/chapter1.pdf.]
Takada, S., 2018b: Support for NWP updates. Report of Numerical Prediction Division. 64, Japan Meteorological Agency, 88-90 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwpreport/64/chapter3.pdf.]
Takada, S., 2018c: The impact of and response to observatory relocation. Report of Numerical Prediction Division. 64, Japan Meteorological Agency, 91-93 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/nwpreport/64/chapter3.pdf.]
Veira, A., R. Hess, S. Trepte, G. Vogt, and B. Reichert, 2017: Model Output Statistics for Point Forecasts at Deutscher Wetterdienst: Current Status and Future Developments. European Conference for Applied Meteorology and Climatology 2017, European Meteorological Society, EMS2017-378. [Available at http://meetingorganizer.copernicus.org/EMS2017/EMS2017-378-1.pdf.]
Wakayama, I., T. Imai, T. Kitamura, and K. Kobayashi, 2020: About estimated weather distribution. Weather service bulletin, 87, Japan Meteorological Agency, 1-18 (in Japanese). [Available at https://www.jma.go.jp/jma/kishou/books/sokkou/87/vol87p001.pdf.]
Wilks, D., 2011: Statistical Methods in the Atmospheric Sciences. 3rd ed. International Geophysics Series, Vol. 100, Academic Press, 704 pp.
Wilson, L. J. and M. Vall´ee, 2002: The Canadian Updateable Model Output Statistics (UMOS) System: Design and Development Tests. Wea. Forecasting, 17, 206–222, doi:10.1175/1520-0434(2002)017<0206:TCUMOS>2.0.CO;2.
Zurndorfer, E. A., J. R. Bocchieri, G. M. Carter, J. P. Dallavalle, D. B. Gilhousen, K. F. Hebenstreit, and D. J. Vercelli, 1979: Trends in comparative verification scores for guidance and local aviation/public weather forecasts. Mon. Wea. Rev., 107, 799-811, doi:10.1175/1520-0493(1979)107<0799:TICVSF>2.0.CO;2.
Downloads
Posted
Submitted: 2023-09-19 04:09:03 UTC
Published: 2023-09-21 02:58:47 UTC
License
Copyright (c) 2023
Takuya Inoue
Tsuyoshi Thomas Sekiyama
Atsushi Kudo
This work is licensed under a Creative Commons Attribution 4.0 International License.
